See the "Relation" table in the specification.
CHIELD is a database of hypotheses, not of definitive proof. The connections should reflect the author's ideas. Causality is very hard to demonstrate, but you can add the type of support (experimental, statistical etc.), and there are options to state that e.g. the experiment did not support the link, or that the authors only propose a correlation (~), or that the authors propose no causal link (/>).
Yes! There are options to state that e.g. the experiment did not support the link, or that the authors only propose a correlation (~), or that the authors propose no causal link (/>).
There doesn't appear to be a standard way of representing interactions in causal graphs. Some have an edge which links a node to another edge (i.e. the node interferes with the effect of another link). That's not possible in CHIELD at the moment. The other option is to have a node whose label is the interaction between two variables (e.g. "content bias * population connectivity). For more discussion, see e.g. VanderWeele & Robins (2007)'s section on sufficient causation.
However, in many cases the issue can be resolved by thinking about the extra steps in between.
If in doubt, keep the structure simple, but add a note in the "Notes" column.
Any peer-reviewed publication. This includes journal papers, conference abstracts and books.
Yes! We want to be able to see overlapping theories, replications and converging evidence. For hypotheses with longer histories, it's often better to just include an up-to-date review or summary of a particular theory (e.g. Dunbar 2004 covers several incremental papers by Dunbar).
The value of papers that test explicit links is that they provide supporting evidence for larger theories. The best solution would be to code the original theory separately (or check if it's already coded), then add the single link from the test, including the type of study.
If a paper has e.g. a cross-cultural statistical test and an experiment, you can add both as seperate links (seperate rows in the table where the entries are mostly the same except for "type" and probably "notes"). Note that multiple links between the same nodes may not always show up in the graphical display.
The github repository managers. If you would like to become a manager, please contact the developers.
Like "shield", but louder.